RTX PRO 6000 Hits 2900 tok/s Running 30B Models
HankYeomans · x · 2026-08-13
A user shared benchmark results for the NVIDIA RTX PRO 6000 GPU. Running the Nemotron-3.5-Lightning-30B-A3B-NVFP4 model, the card achieved an aggregated throughput of 2900 tok/s and secured 8.1M of KV cache. The author noted that while the card is expensive, it is worth the price.
Related event: RTX PRO 6000 Hits 2900 tok/s in 30B Model Local Inference Test(2 posts)→
More from Infra
- Heron Power Invests Over $100M in US Factory for AI Datacenter Grid Tech — espricewright · 2026-08-13
- SK Hynix, Samsung, and Micron Reportedly Sell Out All 2027 HBM Capacity — Beth_Kindig · 2026-08-13
- NVIDIA Releases AI Tokenomics Guide: Turning Compute into Revenue — nvidia · 2026-08-13
- NVIDIA Defines the AI Era: AI Factories as New Infrastructure, Tokens as New Commodities — nvidia · 2026-08-13
- Kubernetes DRA Reaches GA: Native GPU Scheduling and Slicing — sloppenheimer · 2026-08-13
- SanDisk Targets 80% Long-Term Gross Margins and Revenue Growth Through 2030 — firstadopter · 2026-08-13